[tylermenezes]

I have read his spreadsheets, so here's my critique. (And I will say that, throwing all the spreadsheets out the window, I still agree broadly with many points to some extent.)

The input numbers (in meta.xlsx) come from imperfect samples of the real world with non-zero error -- often quite a bit higher than 0 -- which is compounded pretty far in some cases. I don't think any person actually can actually evaluate these studies, which is a major problem with this entire idea and is a fundamental problem with any overly-quantitative solutions to problems.

[Someone]

There also is (FTA):

“I took the countless papers that start with the standard return estimates and tweak them with one novel complication. Then I merged all the tweaks that seemed convincing to me* to get final policy-relevant numbers”

So, the model was adjusted from tweaks created by one person, and then those tweaks were evaluated by the same person, not on objective criteria, but on what seemed right.

I also guess the author at the start had ideas about the quality of each of the “countless papers that start with the standard return estimates” and used them to pick those he included in the model.

I don’t see how that could ever lead to an objective model, even if the author started this exercise unbiased.

[IshKebab]

Yeah I wouldn't look at his numbers for the simple reason that it's a predictive model that's presumably chock full of fudge factors and educated guesses. It's pretty trivial to tweak it (he even admits he did that) to get the result you want. And he clearly has a result that he wants.

Well, that and the idea that education is just to make money... I'm sure there are dozens of beneficial returns he left out, like more educated people voting for parties that are less likely to screw them over.